Can passive repeater concept work for GSM mobile phones?

Steve Jenkin sjenkin at canb.auug.org.au
Sat Sep 4 02:52:30 GMT 2004


On Fri, 2004-09-03 at 14:27, Chris Young wrote:
> Hi,
<snip>

> Can such a concept work for phones - ie roof antenna, cable down into 
> smaller antenna inside house?  I heard active / amplifiers break the 
> law, so passive should be ok. 

Unlikely you'll get what you want, BUT 'RF' is a tricky thing :-)
Will cost you <$100 to rig up a temporary test. [High-gain outside
aerial, co-ax, omni-directional internal aerial.  Don't know how you
impedance match to limit SWR losses (Standing Wave Reflection?)]
(Someone like Carl Makin would know)

Amplifiers, per se, are not illegal - just look at all the mast-head
amps on TV antennas in fringe areas.  They receive, not transmit...
[The same people that do this work would be worth a call, it could be a
'solved problem']

> I can understand if reception is improved.  But what about transmission, 
> ie 'talk'.  Does it work reverse - will signal from phone find the 
> antenna outside, as the path of least resistance?  Otherwise it will be 
> listen-good but talk-bad.

Reception will probably be fine: you've got a fixed transmitter and can
point a high-gain antenna at it, then pipe this power down to an inside
aerial.  I've heard of a guy in a difficult country location (in a
mountain valley) who used two high-gain antennas back-to-back to get TV
reception

Like you anticipate, transmitting back to the base station could be
problematic - you're transmitting from a omni-directional aerial to
another low-gain antenna & piping the power (that's min 3db loss there)
up to the high-gain directional antenna... That first link is very very
lossy.  Standing close to the inside aerial may just work well enough
for you - again, easy to test with a cheap testbed.  GSM phone output
_is_ high: 5 watt max, and that could be enough to get you connection.
[Having to go 'high power' will reduce talk-time battery endurance
considerably]

GSM works @ 900Mhz, well above VHF (TV) frequencies.  It is more
'straight-line' than TV (and less than the 2.4Ghz & 5Ghz of 802.11a&b). 
BUT, at these high-frequencies and modulation styles, signal strengths
in the micro-watt range (-30dbm?) are all that's needed. (anyone know
the numbers for sure?)

A couple more comments in-line below...

Hell, if things are really bad and there's a few of you down this
particular hole, you could get your own 'micro-base-station' [not sure
the correct name].
Not sure how you get Telstra/Optus to put one in, or if you'd be
expected to contribute some. [Remember Senator Alston got one :-) ]

A second-removed family member of mine has done installs of these things
all over Sydney.  Probably $50-100k to install. (Anyone know for sure??)

HTH
sj

> 
> Thanks
> Chris Y.
> 
> >Hi Tony, Robert and group,

<snip>

> >The best example of a passive repeater that I've seen, is on a ridgeline
> >above the town of Karratha, in the Pilbara region of WA.  Two grid pack
> >dishes are mounted back to back, interconnected only by a section of
> >waveguide.  No electronics on site, no solar panels, no 240V.

Telecom/Telstra use this arrangement often.  It relies on:

- Beams & very high-gain aerials for point-to-point, not
broadcast/mobile comms.
- Symmetry of stations: Both use directional aerials & identical xmit
power
- Collimated transmit beam (the spot is not much larger than the
receiving dish at these close ranges & won't wander much with
atmospherics)
- modest power amplifiers (and they can increase power if they need)
- High frequencies, low interference


> >I have also used the passive principle to provide multi-channel VHF radio
> >coverage underground over a leaky feeder system, where there was heaps of
> >signal outside, and none in the underground structure.

This is what they do in road tunnels in Sydney for AM/FM radio _and_
mobile phones...  In great company :-)
I'm not sure if they've done it in train tunnels around the CBD there as
well.

And note the 'leaky co-ax' method works with mobile phone base-stations.
As usual, the trouble is getting that base station connected back into
the trunk network...  That's a landline or microwave link, which you'd
use in preference to the mobile phone if you had the choice...


<snip>
> >Hope this helps,
> >Chris vk6kch
> >
-- 
Steve Jenkin, Unix Sys Admin
0412 786 915 (+61 412 786 915)
PO Box 48, Kippax ACT 2615, AUSTRALIA



More information about the wireless mailing list